kennethashley/bittensor-skill

Overview

This skill lets you query the Bittensor decentralized AI network using the Python SDK and CLI fallbacks. It returns structured JSON about metagraphs, subnets, validators, emissions, staking, and wallet status. Use it to inspect TAO flows, validator rankings, subnet details, and staking activity across mainnet or testnet.

How this skill works

The skill runs Python scripts that call the Bittensor SDK and outputs standardized JSON objects describing netuid, subnet name, neuron counts, top validators, emissions, and stake. For operations not covered by scripts it falls back to the btcli commands for wallet balances, stakes, and subnet registrations. You can target networks (finney/mainnet or testnet) and filter results like top N validators or a specific netuid.

When to use it

• When you need the top validators on a specific subnet and their emissions.
• When you want a list of all subnets with their netuid and basic stats.
• When checking staking information or wallet balances related to TAO.
• When auditing subnet emissions or validating validator rankings.
• When building dashboards or automation that consumes JSON network data.

Best practices

• Specify --network (finney or test) to avoid ambiguous results between mainnet and testnet.
• Use --top N to limit output and reduce bandwidth for large metagraphs.
• Parse the returned JSON instead of scraping text to ensure stable automation.
• Combine SDK queries with btcli fallbacks for wallet or stake operations not exposed by scripts.
• Cache metagraph responses for short intervals to reduce repeated network calls.

Example use cases

• Fetch the top 5 validators on netuid 8 to report emissions and stake for a monthly audit.
• List all subnets to populate a UI selector with netuid, name, and neuron counts.
• Query a subnet metagraph to compute total emission by summing top_validators.emission.
• Check wallet balance and active stakes before submitting a stake or delegation.
• Run regular checks on validator rankings to detect sudden drops or suspicious behavior.